Strategic brief 01Insurance · Q2 2026

Where Insurance Buyers Form Consideration Sets in 2026

Why most brokers are absent from the moment of choice — and the strategic shift required to compete.

Chris Gardner — Digital Marketing Consultant, USA TODAY NETWORK | LOCALiQ. Founder, Signostic. Windsor, ON. Published 21 April 2026 · Reading time ~9 min · About · LinkedIn · Email

The argument

The Quality Score work is real, the conversion-rate work is real, and they are both happening too late. Insurance buyers form their consideration set before they ever reach a paid search funnel — in the AI answer surface, the peer review graph, and the category entry points buyers do not announce. Most brokers compete on the activation half of the budget; the brand half determines whether they were in the consideration set in the first place.

The position

The auction is not where the decision was made.

For roughly two decades, the dominant model of insurance digital marketing has been a paid search auction question: bid on the keyword, optimise the Quality Score, fix the landing page, lower the cost per quote. Every operator-grade discipline in the category — including most of what Signostic audits — lives inside that question. The work is necessary. It is also incomplete.

The decision a buyer makes when they type commercial general liability quote into Google has, in most cases, already been narrowed to a list of two or three brokers they are willing to consider. That narrowing did not happen on the search engine results page. It happened weeks or months earlier, in moments most marketing dashboards cannot see — a renewal notification, a closing date, a conversation at a chamber meeting, a recommendation surfaced inside ChatGPT or Perplexity, a referral on a community Facebook group, a YouTube explainer about umbrella coverage. The auction does not create the consideration set. It harvests it.

This is not a new framework. Byron Sharp and Jenni Romaniuk have made the case for fifteen years that brand growth is driven by mental availability — the probability that a brand comes to mind in a buying situation — far more than by any single touchpoint at the moment of purchase. What is new is the velocity of the change in where mental availability is being constructed. Insurance, the textbook YMYL category, is now being mediated through layers that did not meaningfully exist three years ago.

The mechanism

Category entry points, then the search.

The framework most useful here is the Category Entry Point (CEP), introduced by Romaniuk in Building Distinctive Brand Assets and elaborated in Better Brand Health. A CEP is the cue that brings a category to mind — a moment, a need, a circumstance. Buyers do not think about insurance most days. They think about insurance when a CEP fires.

For an SMB and mid-market insurance buyer, the CEP set looks roughly like this:

Renewal trigger Annual policy renewal date. The single highest-volume CEP for personal lines and most commercial books. Triggers a 30–60 day shopping window in roughly 18–25% of policyholders, per industry retention benchmarks.

Life event House closing, new vehicle, marriage, child, business formation. Each life event triggers a category-entry moment that is durable, intent-rich, and almost always preceded by category research before any provider is contacted.

Claim event Recent claim experience — theirs or a peer’s. A bad claim experience opens the consideration set whether or not it was the policyholder’s. Peer-shared claim stories drive a meaningful share of switching consideration.

Audit event Business audit, lender requirement, board mandate. Particularly relevant in commercial lines. Triggers a structured comparison process that almost always begins with research, not direct contact.

Rate shock Premium increase — theirs or a comparable peer’s. The most volatile CEP because it is largely outside broker control but produces the most switch-driven consideration.

Each of these CEPs creates a research moment. The research moment is where the consideration set is formed. The Google paid auction sits at the end of that research moment, not at the beginning of it.

The mediation layer

Where the research moment actually happens.

The research moment in 2026 is not a single channel. It is a sequence that increasingly bypasses traditional paid search at the front end. A typical research moment for a small business owner facing a commercial general liability renewal looks something like this:

The owner notices a renewal email. They open ChatGPT or Perplexity and ask, in conversational form, “What should I look for when renewing commercial general liability?” The AI engine returns a paragraph answer with two or three named brokers cited, plus an industry body or two. The owner clicks one of those citations. They land on a broker site, scan for credibility signals — named author, dated content, regulatory citations, testimonials — and form a first impression. They then go to a peer channel: a LinkedIn post, a community Facebook group for their industry, a Reddit thread, a Google review trail. They cross-reference. By the time they type a category keyword into Google paid search, they are no longer comparing “all brokers in my city.” They are comparing the two or three brokers their research moment surfaced.

Three things are happening in this sequence that the operator-grade audit cannot see. First, AI answer engines are now a consideration-set channel, not just a discovery channel — if you are not cited there, you are not in the set. Second, the peer channels are providing the trust signal that the broker site itself used to provide — the broker site is now being read after peer validation, not before. Third, the keyword search is the harvest, not the hunt — the buyer arrives at the auction already partially decided.

The implication for a marketing director or VP is straightforward. If your spend is concentrated downstream of the research moment, you are paying full freight to convert buyers who already have a strong preference for someone else. The lift available from upstream investment — brand-building work, AI search visibility, peer-channel presence — is structurally larger than the lift available from a further point of CPC optimisation.

The investment frame

The 60:40 split, in a regulated category.

The Binet and Field central tendency, derived from the IPA Effectiveness Databank across more than a thousand cases, is that the optimal long-run split between brand-building and sales-activation investment lands near 60% brand to 40% activation. The figure is a central tendency, not a prescription. It moves with category, brand maturity, and growth objective. In regulated YMYL categories like insurance, where buyer trust must be constructed before any transaction is plausible, the brand share typically rises rather than falls. Media in Focus (2017) and Romaniuk’s subsequent work both put financial services and insurance in the 65:35 to 70:30 range for mature brands.

Most SMB and mid-market insurance brokers spend nothing close to that ratio. The realistic distribution across the books we audit looks closer to 90:10 or 95:5 toward activation. The brand half is not absent because it is unaffordable — many of the channels that build mental availability in 2026 cost less per dollar of effect than the marginal CPC dollar — it is absent because the operator dashboard cannot measure it on the same time horizon as a paid auction.

The diagnostic question for the VP

Of every dollar your book spent last quarter, what share went to construction of mental availability for the next 12 to 24 months — and what share went to harvest of consideration that was already formed? If the answer to the first half is below 25%, the consideration-set problem is structural, not tactical.

The brokers who outperform their category over a five-year window almost always show a higher brand-share allocation than their peers. Sharp’s work on penetration as the dominant growth lever is consistent: you grow by acquiring new buyers from the much larger pool of category buyers who do not currently use you, and you reach those buyers via the channels that build mental availability before the CEP fires. Activation harvests; it does not build.

The convergence

AEO and brand are the same play.

The most useful conceptual move available to an insurance VP in 2026 is to stop treating AI search visibility and brand investment as separate disciplines. They are not. They are the same investment in mental availability, expressed across different surfaces.

When a buyer asks ChatGPT or Perplexity for a recommendation, the AI engine returns the brokers it has the strongest authority signals for: named authors with credentials, dated content, regulatory citations, third-party recognition, structured data that resolves cleanly into a citation. The same authority signals that drive Answer Engine Optimization are the signals that drive what Romaniuk would call distinctive brand assets in the digital surface — consistent authorship, consistent identity, consistent presence in the category conversation. The technical work of E-E-A-T compliance and the strategic work of brand-asset building are doing the same job for the buyer and for the AI engine alike.

The implication: a brokerage that invests in named-author content, peer-channel presence, and the technical scaffolding that makes that content extractable by AI engines is making a single investment that compounds across both the human research moment and the AI-mediated one. A brokerage that invests only in PPC optimisation is buying the harvest of someone else’s mental-availability work.

The actions

Three moves a VP can make this quarter.

If the diagnosis lands, three moves are immediately available. None of them require pausing the activation work; all of them sit upstream of it.

Audit your AI citation rate against your top three competitors.

Run 30 to 50 category-relevant questions in ChatGPT, Perplexity, and Google AI Overviews. Log which brokers are cited and at what frequency. The output is a defensible mental-availability proxy in the AI-mediated surface, and it is repeatable quarter over quarter. This is the single most useful diagnostic the AEO discipline produces; it costs an analyst day and reframes the strategy conversation immediately.

Reallocate at least 15–20% of next quarter’s spend to brand-building work mapped to your top CEPs.

Not all 60%. Not yet. A directional reallocation of 15 to 20 points of share, mapped specifically to the renewal trigger and the life-event CEP set, is the largest move that is plausible inside one quarter and large enough to produce measurable mental-availability lift over the next 12 months. The brand half can sit in named-author content, longer-form video for the YouTube and CTV surface, and peer-channel investment — not in the abstract category-creative work that has historically priced out small books.

Stand up a single named author as the public face of the book.

The most efficient brand-asset investment available to an SMB or mid-market broker in 2026 is to choose one principal — founder, owner, senior advisor — and make them the named, credentialed, dated voice of every published page on the brokerage’s website. Person schema, sameAs links to LinkedIn and the regulatory registry, byline visibility, and one published piece per month. This single move closes most of the E-E-A-T gap, supplies the AI surface with the authorship signal it requires, and gives the peer channel a recognisable name to attach trust to. It takes an afternoon to set up the schema and a sustained year to compound the authorship signal.

None of these moves replace the activation work. They make the activation work cheaper. A book whose mental availability is healthy buys clicks at a discount; a book whose mental availability is empty pays a premium for every quote it converts. The strategic work is the cost-reduction work, performed one layer upstream.

Primary sources

Where this brief comes from.

Referenced in this brief

Sharp, B. How Brands Grow: What Marketers Don’t Know. Oxford University Press, 2010.
Romaniuk, J. & Sharp, B. How Brands Grow Part 2: Including Emerging Markets, Services, Durables, New and Luxury Brands. Oxford University Press, 2016.
Romaniuk, J. Better Brand Health: Measures and Metrics for a How Brands Grow World. Oxford University Press, 2023.
Binet, L. & Field, P. The Long and the Short of It: Balancing Short and Long-Term Marketing Strategies. IPA, 2013.
Binet, L. & Field, P. Media in Focus: Marketing Effectiveness in the Digital Era. IPA, 2017.
Pew Research Center. Search engine use and the rise of generative AI. 2025 survey data on US search behaviour.
Insurance Bureau of Canada. Industry trends and consumer research. 2025 retention and switching benchmarks.
Search Engine Land. AI Overview citation source analysis. 2025 study on citation-source ranking distribution.

Talk to the author

If a brief surfaces a question for your book — or you want to run the AI citation diagnostic on your top competitors — Chris Gardner reads every inquiry personally. Findings translate into strategy — execution runs through LocaliQ when you’re ready.